Revista Română de Materiale / Romanian Journal of Materials 2015, 45 (2), 105 – 116 105
PROPRIETĂŢILE ŞI CARACTERISTICILE MICROSTRUCTURALE ALE UNOR LIANŢI MICŞTI CU ZGURĂ, ACTIVAŢI ALCALIN
PROPERTIES AND MICROSTRUCTURAL CHARACTERISTICS OF ALKALI-ACTIVATED SLAG-BLENDED CEMENTS
N.R. RAKHIMOVA∗, Kazan State University of Architecture and Engineering, Department of Building Materials
420043, Russian Federation, Kazan, Zelenaya Str. 1
The effects of ground used sand (GUS), ground fly ash (GFA) (class F), and microsilica (MS) on water requirement, setting
time, and compressive and bending strength development of alkali-activated slag-blended cements (AASBC) were studied. Siliceous blending materials were found to be able to replace up to 50% of ground granulated blast furnace slag (GGBFS) and contributed to up to 90% improvement of strength of AASBCs, The granulation for GUS, granulation, curing conditions, and basicity of GGBFS for GFA, and curing conditions for MS have effects on the development of the properties of the AASBCs.
Keywords: Blast furnace slag, alkalis, activators, binding composite materials, setting time, mechanical properties 1. Introduction
Sustainable development in the construction
industry requires effective blended cements and materials based on these blended cements. The introduction of supplementary cementitious materials (SCM) in traditional and alternative cements, in addition to solving imperious ecological problems, also improves the performance capabilities of the cement.
Extensive literature on the research and application of blended ordinary Portland cements (OPC) is available; however, the potential of the combination of various types of SCMs and OPC are yet to be exhaustively studied and reported. Further, sustainability in this field is associated with an increasing use of known SCMs, development and use of new SCMs, and the development and use of different clinker types. This requires a scientific approach to understand the reactions and performance of such materials combination [1].
SCMs are classified into two types, inert and active, which are the most frequently used classifications in building materials science. It is usually to name mineral powders that react with OPC and water as reactive SCMs, and those that do not form hydration products with binding properties as inert SCMs or simply fillers because of the insignificant interaction of the cement paste with such materials. This distinction is clearly relative because all types of mineral powders have some degree of effect on the structure and properties of mixed binders and therefore, in addition to being active, they are also multifunctionally active, differing only in the mechanism of influence on the structure and
properties of the filled binders. Therefore, it seems reasonable to classify mineral blending materials based on whether they are “chemically active” i.e. they form hydration products with cementitious properties or are “physically active” i.e. they do not form hydration products. The “physically active” mineral blending materials affect the physical structure and properties of blended binders, i.e. they are “physically active and reactive” supplementary materials and they combine both “filler” and “modifier” effects [2].
This approach is very reasonable for alkali-activated slag-blended cements (AASBC). The substitution and modifications represent the most promising trend for alkali-activated slag cements (AASC). Alkali activation allows for effective interaction between the AASC paste, fillers, and modifiers and enhances the compatibility with mineral blending materials of various compositions and structures. Hence, a much wider range of various mineral materials are usable with AASBCs than with blended OPCs. Therefore, the properties of AASCs are dependent on many more factors than OPCs. In addition to the influencing factors common to all powdery binders, i.e. chemical and mineralogical composition, fineness, water/binder ratio, and curing conditions, the influencing factors related to the alkaline component, its composition, and concentration, have a significant effect on the properties and structure of the AASCs.
Influencing factors of blending materials for AASBC are similarly to those for OPCs, and they include composition, fineness, and the amount of glassy phase. Due to the diverse range of SCMs used, generic relations between the composition,
∗ Autor corespondent/Corresponding author, E-mail: [email protected]
106 N.R. Rakhimova / Properties and microstructural characteristics of alkali-activated slag-blended cements
particle size, and exposure conditions, such as temperature or relative humidity become increasingly crucial [3].
Some of the main influencing factors controlling AASBC and their significance have been investigated. In this aspect, alkali-activated slag-fly ash systems have been most extensively researched.
Smith and Osborne [4] were the first which reported on alkali-activated slag-fly ash systems. A combination of 60% slag and 40% fly ash with 7% NaOH (relative to the combined mass of slag and fly ash) showed good early mechanical strengths and a small strength gain after 28 days.
Tang [5] investigated the effect of slag/fly ash ratio, water/binder ratio and activator (waterglass and NaOH) dosage on the strength development of alkali-activated slag-fly ash cements using the factoral design method. He found that these factors affected the strength of the cement in the following order: slag/fly ash ratio > water/binder ratio > waterglass dosage > NaOH dosage.
The activation of fly ash/slag pastes with NaOH solutions have been studied by Puertas [6]. The ratio of fly ash/slag and the activator concentration were found to be significant influence factors. The influence of curing temperature on the development of the strength of the pastes was lower than the contribution of other factors.
The effect of the fineness of ground fly ash with a specific surface area 400 m2/kg on the properties of alkali-activated mortars was studied by Lu C. [7]. Mixed AASC showed a 28-day strength of 76 MPa for a fly ash content of 60%. When non-ground fly ashes were used, the 28-day strength was only approximately 12.9–16.2 MPa.
Partial replacement of the ground granulated blast furnace slag GGBFS by fly-ash gives positive results in improvement of performance capabilities and substitution of slag. Proper slag/fly ash ratio was found to have an effect of a mutual complement to the superiority supplement [8]. Alkali-activated slag-fly ash cements showed very good resistance to acidic, sulphate, and seawater attacks, and corrosion resistance increased with increase in the fly ash replacement [9], which also improved the mechanical strength and pore structure [8, 10-12].
Many studies have demonstrated the complex microstructure and chemical composition of alkali-activated slag-fly ash cements binder gels. Shi and Day [13] and Puertas and Fernandez-Jimenez [14, 15] reported that the main hydration products of alkali-activated slag-fly ash cements were C–S–H and alkali aluminosilicate hydrate gels. Yang found that the microstructure and chemical composition of such blended binders fall between those of the aluminosilicate gel formed in the silicate-activated fly ash and those of calcium
silicate hydrate gel formed in silicate-activated granulated blast furnace slag (GBFS) [16].
Thus, the microstructure, composition, properties, slag/fly ash ratio, water/binder ratio, alkali activator dosage, grinding fineness up to 400 m2/kg, and curing temperature have been investigated as influencing factors for alkali-activated slag-fly ash cements. The influencing factors and their significance in AASBC with “physically active” and “chemically active” blending materials, to the best of our knowledge, have not been reported. In addition, the relevance of the above mentioned influencing factors and the fineness of grinding up to specific surface areas of 800 m2/kg, GGBFS basicity, and pre-hardening time on the property evolution and their dependence on the activity of the blending materials is not completely known. In the present study, we have studied the effect of the typically physically active blending materials, such as silica sand, which is physically active and reactive, fly ash and reactive mineral additions, and microsilica on the properties of fresh and hardened pastes of AASBCs. In addition, we have also attempted to determine the significance of the influencing factors on the property evolution of AASBC. Further, we have attempted to describe the main structural characteristics of AASBCs depending on the activity type of the blending materials.
2. Experimental
The starting materials were ground granulated blast furnace slags (GGBFS) from Orsko-Khalilovsky (GGBFS 1) and Chelyabinsky (GGBFS 2). Specific surface areas (Ssp) of the GGBFS were 300 m2/kg. As the blending materials were used industrial by-products of siliceous composition, which in accordance with the proposed classification, can be considered as physically active, physically active and reactive, and reactive i.e. ground used sand (GUS), ground fly ash (GFA), and microsilica MS (Table 1, 2). The GBFSs, used sand (US), and fly ash (FA) were ground in the laboratory planetary mill.
The alkaline activation of the mixed cement was carried out using a commercial sodium carbonate solution. The addition of the alkali activator was adjusted to 5 wt% (by Na2O) of the blended cement.
The AASBC cement pastes were prepared in cubic moulds (2 × 2 × 2 cm3), cement-sand mortars in prismatic moulds (4 × 4 × 16 cm3). The compressive strength of AASBC was measured after 1, 3, 7, 28, and 365 days of storage under normal conditions (room temperature, humidity 95–100%) and after steam curing regime for 4 + 3 + 6 + 3 h (four hours of pre-hardening time at room temperature, three hours of temperature rise time, six hours of constant temperature 90–95 °C time, and three hours of temperature decrease time).
N. R. Rakhimo
Starting material GGBFS GGBFS FA (class F
Blendin
GG
Ble
ndin
g m
ater
ial
-
GU
S
GFA
The was evaluawhich is thspecimen blending mspecimen b
3. Results
First,was studiedgrinding theand for US
Fig. 1 - Relatio
time o
ova / Proprietă
SiO2 1 40.02 2 37.49
F) 58.76-62.1
ng material
GUS GFA
MS
The cha
Con
tent
of b
lend
ing
mat
eria
l in
AA
SB
C, %
Nor
mal
cons
iste
ncy
0 220 240 260 220 240 260 3
effectivenesated by the
he ratio of thobtained bymaterials tobased on GB
and discuss
the grindabd. Figure 1 she GBFS to and FA to ac
onships betweenof grinding.
ăţile şi caracteris
CaO A42.02 36.22 16.12-8.64
11
Content of Si
90.1-96.58.76-62
93.9-94.
aracteristics of th
200
Nor
mal
con
sist
ency
, %
Settingm
initial
24.9 1-20 24.9 1-10 24.8 1-50 24.8 3-00 27,2 2-00 29,4 3-50 31,6 6-40
s of the blee effectivenehe 28-days y replacing o that of FS.
sion
bility of the sthows the timachieve a Schieve a Ssp
n Ssp of the star
sticile microstru
Chemical com
Al2O3 MgO8.22 6.26
11.58 8.615.46-
19.31 1.01-2.78
CharacteristiO2, % Co
5 .1
7
he fresh paste o
g time, hour-minute
final
3-50 3-40 4-20 5-50 4-40 6-40
10-20
ending materess factor (strength of the GBFS the refere
tarting materme necessarySsp of 300m2
of 800m2/kg
rting materials a
cturale ale unor
mposition of GBFComponent (
MnO F0.34 <0.50 00.09-0.55
79
cs of the blendiontent of amorp
phase, %0
65
100
of the AASBCs wSsp of blending
5N
orm
al c
onsi
sten
cy,
%
Se
ini
24.9 1-25.3 1-25.7 2-26.1 3-27,4 3-29,9 8-32,3
rials (Fe), the by
nce
rials y for 2/kg .
and
3
fr
nais5rerewisue
h
bawAaaTsthin
r lianţi micşti cu
FSs and FA (mass % as oxid
Fe2O3 TiO2 <0.1 0.36 0.16 1.80 7.36-9.33
0.23-0.95
ng materials phous
15
with GFA and Gg materials, m2/k500 tting time, hour-
minute
itial final
-20 3-50-20 3-50-20 5-00-50 7-00-00 5-30-20 11-30- -
3.1. PhysicalThe in
resh paste.
The wot increase wdded (AASB
s increased u00 and 800 equirement espectively.
with GUS cons slightly redp to 60%, duxtended by 2
Influenchardened pas
Figureslending matnd its develo
with GUS (FAASBC is ba
nd is activalkali activato
The compresamples increhe GGBFSnfluencing fa
zgură, activaţi a
de) Na2O K0.44 00.64 00.35-0.97
01
Ssp, m2/kg
200-800 200-800
000 - 25 000
GUS as blendingkg
-
Nor
mal
con
sist
ency
, %
24.9 25.5 26.1 26.7 28,1
0 31,3 34,5
lly active GUfluence of t
water requirewhen GUS w
BC (GUS)) anup to 60%. Tm2/kg up to from 24.9 The setting
ntent of up touced. With iue to dilution2–3 times (T
ce of the ste s 2–4 showterials on thopment. Rep
Figure 2) is ased on the ated by a mor i.e. a sodssive strengteases with inand of curictor respons
alcalin
K2O P2O5 0.66 0.04 0.95 0.01 .56-.05
0.05-0.09
Mineral co
quaquartz – 7-1
– 15-1Fe-spinel –
than qua
g materials
800 Setting time, h
minute
initial fi
1-20 31-40 42-50 54-40 83-50 69-40 14
-
US the GUS on
ement of thewith a Ssp of nd when the
The introducti60% increas
to 26.1 g time of AAo 20% does nncrease in t
n effect, the sable 3).
GUS on
w the influehe compressplacement o
beneficial, neutral or amildly activedium carbonth of the AAncrease in thng temperat
sible for the m
10
Table 1
SO3 1.45 2.00 0.17-0.98
Table 2
mposition
artz 1%, mullite 18%, – no more 5%
artz Table
hour-
nal
-50 -10 -30 -20 -20
4-20 -
n the AASBC
e AASC doef 200 m2/kg iGUS conten
ion of GUS oses the wateand 26.7%
ASBC (GUSnot change ohe content tsetting time i
the AASBC
ence of thsive strengtf the GGBFSalthough thcidic GGBFS
e non-silicatnate solutionASBC (GUShe basicity oture. A maimaximum
07
3
C
es s
nt of er %, S) or o s
C
e th S e S e
n. S) of n
108
Fig.2 - The va
(GUS) wGGBFSc) GGB
ariation of the with Ssp, conten
S1, normal condBFS2, steam cur
N.R. Rakhimo
2a
2b
2c
compressive snt of GUS, and cditions, b) GGBring.
ova / Properties
trength of AAScuring conditionFS1, steam cut
and microstruc
SBCs s: a) tting,
p(GFmSapaegaagTopinost(ethGac
amdinGsthostdstthas(F
F
ctural characteri ossible GUSGUS) is the
Figure 2, thmanifested Ssp(AASBC-2
ttributed to sarticle redmorphizationnergy. Secorain-size comount of pand is moreround from 2
The grain-sizef 300 m2/kg articles with
ntroduced [1ptimum distrength of effective conhe dispersionGGBFS up to
t the level oontent of GU
The pplicability
materials is tevelopment.
ndicate that GUS (Ssp = 5trength. Saardened 3–7f the referetrengths of ays are verytrengths of he referencefter one yeample and Figure 4).
ig. 3. - DeveloAASBC28 day
istics of alkali-ac
S content ande fineness ohe positive when Ssp
2,3,5-8) by several reas
duction lean and incondly, there imposition ofarticles smal than 30% 200 to 500-8e distributionis not optim
h sizes les18]. Introdupersion incrthe harden
ntent of GUSn of GUS ano 50%, whileof the refereUS = 0–50%)important
of suppltheir effect o. The resulthe substitut500 m2/kg) ramples of 7 days show ence samplethe blended
y close to ththe samples
e value. An iar is 10.8–3
18.4–48%
opment of theCs in comparisys (a) and 360 d
ctivated slag-ble
d its influencof GUS. On
influence exceeds
1.6 times. sons. First, tds to su
creases theis an improvf the mixed ller than 5 µ[17] when
800 m2/kg (bn of the GGBmum and imsser than ction of GUreases the ned paste
S = 0–30%) dnd allows to se maintainingence sample). characteristementary
on the long-tts in Figuretion of GGBreduces the
the blendstrengths lo
es by up td cements hat of the res after 360 increment in36.7% for t
for blend
e compressive on to the refer
days (b).
ended cements
ce on AASBCthe basis oof GUS i
the GGBFSThis can bthe GUS sizurface layee superficiavement in thcement. Th
µm increasethe GUS i
by 4.5 timesBFS at an Ss
mproves whe20 µm ar
US with thcompressivup to 14%
depending osubstitute thg the strengte (permissibl
tic of thcementitiou
term strengtes 3 and FS with 30%compressiv
ed cementower than thato 20%. Thhardened 2
eference, andays excee
n the strengthe referenc
ded cement
strength of thence up to afte
C of s S e e
er al e e
es s ). sp n
re e e
% n e h e
e s h 4
% e ts at e 8 d d h e ts
he er
N. R. Rakhimo
Fig. 4. - Incremdepenadmix
Acco
[20], Deja [characterizetransitional AASC. Our
Ty
AA
AA
AA
ova / Proprietă
ment in the strending on the xture.
ording to Gl21], Xu et aed by dens
zone betwresults seem
Type o
AASBCAASB
AASBC
ype of AASBC
Reference ASBC(GUS)-0
ASBC(MS)-0
ASBC(GFA)-0
ăţile şi caracteris
ngth after one ycuring condit
ukhovsky [1l. [22] AASC
se and unifoween the am to indicate
Fig. 5 - Sca
of AASBC
C(GUS)-0 C(MS)-0
C(GFA)-0
Strength cha
Comp
sticile microstru
year of the AASions and type
19], Shi et C materials aorm interfacggregate anthat the
anning electron
Experimental cGGBFS
GGBFS1 GGBFS1 GGBFS1
racteristics of th
pressive strength(MPa)
44.2
43.9
58.4
57.6
cturale ale unor
SBCs e of
al. are ial nd
stincdpp
e>G
mththinst
3
fr
th
image of the ha
composition of th
Admixture
GUS MS
GFA
he AASC and AA
h Bendin
r lianţi micşti cu
trengtheningn AABCsontributesevelopment.hase of the article, and i
The efffficiency of t
> (GGBFS baGUS = 0.9–1
The comortars AAShe maintenahe level of thn the bendintrength/bend
3.2. Physical
The inresh paste
The adhe water requ
ardened AASBC
he AASBCs Conten
admixtur303
30
ASBCs mortars
ng strength (MP
7.1
6.2
10.1
10.7
zgură, activaţi a
g of the inte(GUS) wito the
. Figure 5 hardened Anterfacial zofect of the inhe GUS folloasicity, curin.14.
ompressive aBC(GUS)-0
ance of the he referenceng strength ading strength
lly active anfluence of t
dition of GFAuirement and
C(GUS) paste.
nt of re, %
Sa
15
s (steam curing)
Pa) Compre
alcalin
erfacial transith time s
long-term shows the
AASC(GUS) ones in betwenfluencing faows the ordeng conditions
and bending(Tables 4 acompressive
e sample andand ratio of
h.
nd reactive Gthe GFA on
A results in ad this increas
Ssp of the admixture
500 000-25000
500
) essive strength/
strength
6.2
7.1
5.8
5.3
10
sitional zone substantially
strength amorphous
paste, GUS een. actors on th
er: Ssp of GUSs, etc.), Fe o
g strengths oand 5) showe strength ad a decreascompressiv
GFA n the AASBC
an increase ise depends
Table 4
Table
/ bending
09
e S of
of w at e e
C
n
5
110
on the amsetting timeSsp of 200, 5by up to 1.5
Influe
hardened pIn ge
basicity andof Na20 5%is added tGGBFS ispolymineralcombine thparticles of blending mdifferent sizlarge sizes and are ablended AAGFA on thon the Ssp, is shown in 200 m2/kg ("physical acincrease in(AASBC-9,10,13,16) pozzolanic
Unlik(GUS), GFAwhich is 1,5when steamThe permiswhen the comparabledepending curing is usof 200 to 50of particles the solubiliinteraction w
Fig.6 - Th
mount and fes of the AA500, and 8005, 3, and 4 tim
ence of thpaste eneral, glassd slow reacti
% compared to GGBFS, s diluted. l- and quare qualities o
glassy GFmaterial (quzes). The GF
in the first additional aASBC(GFA)ce strength ocuring condFigure 6. W
(AASBC-11,1ctivity" of n the Ssp fro12,15) and facilitates
reaction of thke physicallyA exhibits ph5 times lowem curing is ussible conten
strength e to that of thon the basic
sed. The grin00 and 800 mof sizes beloty of GFA with the mine
N.R. Rakhimo
he dependence
fineness of ASBC(GFA) w0 m2/kg was mes, respect
he GFA on
sy class F Gon to alkali ato GGBFS [2the main
However, rter-phase Gof the reactivFA) and phartz and muFA particles
stage procealuminosilicacement. Theof AASBC(Gditions, and G
When a GFA w14) is
GFA predom 200 m2/k
800 respectthe appea
he GFA. y active blenhysical activer than that osed (AASBC
nt of GFA ofof blende
he reference city of GGBFnding of FA m2/kg increaow 5 µm by min the basiceral matrix of
ova / Properties
of the strength
the GFA. Twhen GFA wused increa
tively (Table
n the AAS
GFA has a at concentra23]. When Gcomponent unequigranuGFA (Tableve additive (hysically acullite particleof medium eeds are fil
ate sources e effect of
GFA) dependGGBFS basiwith a low Ss
used, dominates. kg to 500 m2
tively (AASBarance of
nding materity at a low of GGBFS, o
C-11) (Figuref Ssp 200 m2
ed cement can reach 5
FS when steto reach an
ases the contmore than 30c medium, af the AASC
and microstruc
of AASCs (GFA
The with sed 3).
SBC
low tion
GFA i.e.
ular, 2) fine
ctive s of and lers
in the
ding icity sp of the An
2/kg BC-the
rials Ssp, only e 6). 2/kg
is 50% eam
Ssp tent 0%, and
[2A(AucatethhaGcsm
pmsoTap
hcpohthoinmstegthdWtio
ctural characteri
A) on the curing
2,23,24]. AnAASC mixed AASBC 9,10p to 62%, during conditilso become
emperatureshe incremenigher and is nd in norma
GFA contentonditions onimilar to th
materials - SCThe pr
hases in GFmineral matrixize at high tef the effect o
These factorsnd basicity roperties of b
Increasours to 16 hontent and are-hardeningf silica in Aardening, w
he thresholdxide from
ntergranular maintains theufficient higemperature,lassy phase
han the disecreases to
With increaseme from 2 tof AASBC (G
istics of alkali-ac
conditions, Ssp
n increment with GFA o
0,12,13,15,16depending onons, and Ssp
es much mothan at ro
nt in strengtup to 62% a
al conditionst of 20-25%n the stren
hat of blendCM. resence of thFA, the increax when the Gemperature iof GFA on ths contribute
and curinblended cemse in the prhours allowsaffects the stg at room teAASBC (GF
when the admd value, a ce
the GGBspace. This
e pH into tgh level. the release
e increases solution of a crucial lev
e in the durao 16 h, the GFA) increase
ctivated slag-ble
, and GFA conte
in the streof Ssp of 506) at contenn the basicitp of GFA (Fiore reactive
oom temperath after steaat GFA contes (28 days) u%. The effength of AASded OPC w
he crystallinase in the soGFA particlesnfluence the
he AASBC(Gas much asg condition
ments. re-hardening s an increasetrength (Figuemperature, FA) is low. mixture contertain amouFS liberate
s calcium oxthe binder With incre
e rate of silicsignificantly GGBFS, an
vel at lower ation of the pGFA content es (Figure 7)
ended cements
ent.
ength of th00–800 m2/knt 20-30% ity of GGBFSgure 6). GFA at elevateature; henceam curing ient of 25-30%up to 17% aect of curinSBC(GFA) iwith siliceou
e and glassolubility in ths decrease i
e dependencGFA) strengths GGBFS Sss affect th
time from e in the GFAure 7). Durin
the solubilitDuring pre
tent is belownt of calciumes into thxide probablsystem at ase of thca from GFAmore rapidl
nd pH valuGFA contenpre-hardenin
and strengt.
e g s
S, A d e, s
% at g s s
sy e n e
h. sp e
2 A g ty e-w m e y a e A y e t. g h
N. R. Rakhimo
Fig. 7 - The
(GGhar
The
(GFA) abrucontent (Fiare steam Thus, the and GFA aGGBFS, pacritical cotransformatparticles ofcaused by content of binder systAASC promglass in GFsimultaneoudepolymerizthe total reaa pH < 8, thwith an incrup to 11 to pH increaseand [Si(OHand is eapolycondenInitially, theand then wdecrease interconnecacid gel [26 OH
OH – Si – OH
OH
OH
HO - - Si – O
OH
ova / Proprietă
e dependence GBFS1, GFA 80rdening time.
strength ofuptly decreagure 6), espcured and appropriate re limited, dearticle size oncentrationtion to orthof hydrated s
the mutual glassy silica
tem of AASmotes the r
FA [25]. The sus processzation [26], waction: (SiO2he solubility orease in the 25 times. T
es because H)6]2- [27]. Oasily conve
nsation evene chain polywith an incof рН v
cted by the 6-28] accordi
OH
H + OH – Si – O
OH
O H
n+2
ăţile şi caracteris
of the stren00 m2/kg, steam
f blended ases at a pecially whewhen the Sconcentratio
etermined byof GFA, cu of silic
osilicic acid-silica. Hypoth
conditionala and pH ofBC. High-alreactivity of solubility of sses of hwhich can b2)n + 2nH2O of silica is 2
pH, the soluhe solubility of the forma
Orthosilicic aerted into an in very dycondensatiocrease in covalue, the formation o
ng to the foll
OH OH –
-H2O
sticile microstru
ngth of AASBCm curing) on the
binder AASparticular G
en the sampSsp >500 m2
ons of GGBy the basicityring conditioca, and -bound collohetically, thisity of threshf the hardenkali mediumaluminosilic
silica consisthydration be described↔ nSi(OH)4× 10-3 mol/L aubility increaof silica at h
ation of Si(Oacid is unstaa polymer dilute solutioon takes plaoncentration
chainsof a polysiliowing reacti
OH OH
– Si – O – Si – O
OH OH
cturale ale unor
C-13 pre-
SBC GFA ples /kg.
BFS y of ons,
its oidal s is hold ning
m of cate ts of and
d by 4. At and
ases high
OH)4 able
by ons. ace, or are con on.
oS8HApti
resinethodcvthccccScinbdlocreinthowFoSccmAa
OH +
r lianţi micşti cu
Undissrthosilicic ac
SiO44- is the m
–11.71 orthH3SiO4
- and wAccording to
olymerizingme accordin
O
HO – Si – OH
OH
The seactive aluignificant a
ntergranular ntrance of a
he disruptionf the liquid ecline. Wheertain level,ery short pehe hardeningolloidal partompressive omparison ontent is
Simultaneousalcium oxide
n the рН. Talance in thepend on th
owering of thontent is loequired for nsufficient fohan the criticcur and the
with the formFigure 6, it ca
f FA to achiSsp of GGBFonsiderable haracteristic
m2/kg is sufficAASBC(GFA)
ctivated by s
+nSi(OH)4 -nH2O
zgură, activaţi a
ociated H4Scid existencemajor form ohosilicic acidwith an ion H[29], monom(condensingg to the follo
H HO –
small GFA uminosilicateamount of spaces of th
an acid oxiden of the acid-
medium ofen the pH o
the silicic aeriod and theg cement pticles of hyd
strength sto the sam
lower thasly, the exe (basic oxidThus, the che liquid phas
he influencinghe рН of the ower than the silicic a
or decrease ical value, pe mixed cem
mation of a han also be oieve a Ssp >
FS is 300 mimprovem
s of GFA-AAcient and is t) based on nsodium carbo
alcalin
SiO4 is the me at pH = 8;of existence d is in equ
H2SiO42- at pH
meric H3SiO4-
g) in a very sowing schem
O O
– Si – O – Si –
OH OH
particles ce glass i
silicic acihe hardeninge (silicon ox-base balancf the ceme
of the solutioacid ions poe intergranuaste is filled
drated silicasignificantly mples in whan the cxtraction frde) promoteshanges of thse of GFA-Ag factors of medium. W
the permissacid formatithe pH at a
polymerizatioment fresh phigh-strengthobserved tha
500 m2/kg, m2/kg, fails
ment in thASC. Hence,the upper limneutral and aonate.
11
major form o; at pH = 12[27]. At pH
uilibrium witH = 11.71–12- is capable ohort period oatic process
O
– O – Si – OH
OH
consisting ointroduce id into th cement. Th
xide) leads tce and the pHnt begins t
on reaches lymerize in lar spaces od with boun. Hence, th
worsens ihich the FAcritical limiom GGBFSs an increashe acid-basAASC systemboth rise an
When the GFAsible amounon, which ia lower leveon ceases taste harden structure. It the grindinalthough thto lead to he strengt, Ssp of ~ 50mit for GFA iacid GGBFS
11
of 2, = h
2. of of .
of a e e o H o a a of d e n A t. S e e m d A nt s el o s n g e a h 0 n
S,
112
GGBFS
GGBFS 1
GGBFS 2
Content oAASB
01357
1020
With compressivintroductionstrength, aGUS (Figurthe strengthof the referyear, AASBand AASBC
Comreactive supermissibleThe resultstrength ofreplacemenin the casesupplementof the mainSsp of the depending o
For pinfluencing the followinGGBFS bas
The AASBC(GFThe additiocompressiv 3.3. Reacti
The iMS h
paste. The up to 20% 19.4% (Tabresults obtaOC-based s
Curing conditions
normal conditions
steam curing
normal conditions
steam curing
of MS in BC, %
0
3 5 7 0 0
regard to ve strengthn of GFA lealthough lessre 3). Howevh of GFA-AArence after 2BC(GFA) is sC(GUS). bining the q
upplementarye and effectivs illustrate f the AASCnt of GGBFSe of blendedts increases n constituenadditives. T
on its Ssp. physically acfactors affe
ng order: (Sssicity) > pre-
strengthFA)-0 mortarsn of 30% of
ve and bendin
ive MS influence of Mhas a plastic
water requidecreases fble 7). Thisained by Kosystems, the
N.R. Rakhimo
up to
g up
u
g u
Ch
Normal consistency,
% 24.9 23
20.8 19.5 19.0 18.8 18.7
the develoh of AASds to a dimin
ser than thatver, by introdASC is nearl28 days agesuperior tha
ualities of thy material, Gve replacem
that GFA C and can S by up to 50d OPC, the e
with increasnt, curing teThe Fe of G
ctive and react the efficiesp of FA, cuhardening tim
h characs are presenGFA to AASng strength o
MS on AASBcizing effect orement for a
from 24.9–25s is in agreeorolev [30]. e introduction
ova / Properties
Ranges of repla
effeco 40-50% (FA S
p to 50% (GFA S
p to 30% (GFA
p to 30% (GFA
aracteristics ofGGBFS1Setting ti
initial
1-20 1-10 0-45 0-45 0-40 0-40 0-45
opment of BC(GFA),nish of the et determinedduction of Gy similar to te and after on the refere
he filler and GFA can be
ment of GGBincreases
be used as% (Table 6).effectiveness
se in the basimperature,
GFA is 0.9–
active GFA,ency of GFAuring conditiome. cteristics nted in TableC increases of the mortar
BC fresh pason AASBC(Madditions of 5.8% and 18ement withAs opposed
n of MS into
and microstruc
acement of GGBRanges of the
ctive Ssp 500-800 m2/k
Ssp =200 m2/kg)
Ssp 800 m2/kg)
Ssp 200 m2/kg)
fresh paste of t
ime, hour-minut
final
3-503-402-502-302-101-502-00
the the arly d of US, that one nce
the e a
BFS. the
s a . As s of icity and
–1.6
the A in ons,
of e 5. the
rs.
ste MS) MS
8.7–the
d to
AaNwM
p
ssta
F
stchA
ctural characteri
BFS by GFA e replacement G
kg)
)
he AASBCs wit
te Noconsis
%
25 23 21 19 19 19 19
AASC doesssume that
Na2CO3 solutwater releaseMS.
The inpaste
MS, dignificant potrength of And Table 4).
ig. 8 - The dep
content For rea
trengtheningonditions, tardening tim
AASBC(MS) b
istics of alkali-ac
GGBFS by GFA p
u(GFA Ssp
u(GFA Ssp
th MS GG
rmal stency, %
S
5.8 3.9 1.5 9.9 9.7 9.5 9.4
not increathe interact
tion binder sye, leading to
nfluence of M
ue to its hositive affec
AASBC(MS)
pendence of the, curing conditio
active MS, thg effect is dhe basicity
me. The varbinders with
ctivated slag-ble
ermissible
-
up to 30% p =500-800 m2/k
-
up to 30% p =500-800 m2/k
GBFS2 Setting time, hou
initial
0-50 0-40 0-30 0-25 0-25 0-25 0-35
ase water dtions in the ystem resultso a plasticiz
MS on AASB
high reactivct on the binders (Fig
e strength of AAons, and GBFS
he optimum determined b
of GGBFSriation of thethe MS cont
ended cements
Table 6
kg)
kg)
Table
ur-minute
final
3-00 2-50 2-30 2-20 2-10 2-10 2-30
demand. WGGBFS-MS
s in the “freezing effect o
BC hardene
vity exerts compressiv
ures 8 and
ASC(MS) on Mbasicity.
addition anby the curinS, and pree strength otent
7
e S-e” of
d
a e 9
S
d g
e-of
N. R. Rakhimo
Fig. 9 - The
AAS resembles binders withThe GGBFSa 37% instrength ofconditions wis 27% wheThe upperAASBC(MSfor binderscarried out (MS) bindeunder steacompressivMPa with 4on GGBFSwith 3% MS
The sfor diminishexhibits strbinders. Foseven daysbinders arerespectivelysample. Aftof AASBC(reference AAASBC(GFstrength in the consideAASBC - (M4.1 % for cu
For hof MS, theconcentratio4% with stecuring condsolubility oencounterenormal coinfluencing MS-AASC conditions time. The va
ova / Proprietă
dependence SBC (MS)-17 on
the gradatioh GFA conteS1 is more bcrease is f MS-AASCwhen GGBFen GGBFS2 wr limit of
S) binders bs based on
under normers based oam, the adve strength in4% MS conte 2, the incre
S content). strengtheninh the dosagerengthening or early age os, the compe higher wiy, in compater 360 days(MS) bindersAASC, the
FA). Howevepercentage f
ered binders MS) cured uuring under shighly reactivere are onons. The effeam curing ditions. This of MS silicad during ste
onditions. Tfactors on tbinders fo
> basicity oalue of Fe of
ăţile şi caracteris
of the strengthn the pre-harden
on of that oent of Ssp = 5basic than Gseen in th binders cuS1 is used awas used (FMS conte
ased on GGGGBFS2 w
mal conditionon GGBFS1dition of Mncrease of uent ) and for ease is up to
g effect of Me of the alkaeffect on th
of hardeningpressive streth up to 1arison with s, the comprs outbalancebinders AASer, the incfrom 28 daysis minimum:nder normal
steam (Figurve additions,ly certain efective AASCand upto 7%is explained
a at higheeam curing tThe significthe compresollows the of GGBFS >
MS is up to
sticile microstru
h of AASBC (ning time.
of AASBC(G500–800 m2
GBFS2. Hene compressured in norand the increigures 8 andnt is 7%
GBFS1 and when curing
ns. For AASB1, when cu
MS leads top to 105% (samples ba
o 60% (98 M
MS can be uali activator. he AASBC(M three days
engths of th05 and 50
the refereressive strenes those of SBC(GUS) a
crement in s to 360 day: about 28 %l conditions ae 4). , as in the ceffective AAC quantities % under nord by the higr temperatuthan that uncance of sive strength
order cu> pre-harden1.9.
cturale ale unor
(MS)
FA) /kg. nce, sive rmal ase
d 9). for 5%
g is BC-ured o a 115 sed
MPa
sed MS
MS) and ese %,
nce ngth the
and the s of
% for and
case ASC
are rmal gher ures nder
the h of ring ning
3
3
cfrareainAc8ininasepthGmfainAinre3
cafoinOthfoztrthregmhemtr
3
FAcastcp
2(GG
r lianţi micşti cu
3.4. Structustructof the
3.4.1. AASBCDespite
hanges the rom the momn alkali activeact with alknd Na2O to
ncrease in cAASC. Introdoncentration%.The incre
ncreases then the initial stre maintainupplementarrosion of throlongs the
he eroded sGUS has a himatrix of AASacilitate the nterfacial zoAASBC (GUSndirectly conesults of AAb).
Studiesement freshctivated slag
ormation of nterfacial traOPC. Also, trhe results oormation of one. Evolutransitional zohe adhesionelease of thradual eros
mineral matrardened palements (Fig
matrix of Aransitional zo
3.4.2. AASBCPolymi
FA causes aAASBC (GFAonsisting of ffects the tages of haomponents article size a
Since G9% - quartzGFA) with an
GGBFS ratio
zgură, activaţi a
ure formatiture types, ae AASBC haC(GUS) binde the low hardening p
ment of the mvator solutionkali, the alkalo (CaO, SiOomparison w
duction of 3n of Na2O beased contee activation atages of hardned over ry alkali amohe GUS pardevelopmenurface of thgh adhesion
SC. In the prfaster form
one during S) hardenednfirmed by thASBC (GUS)
s on the inteh paste andg mortars [2
the strongansitional zoraces of che
of our studiea developin
tion of thisone over timn strength, he surface Gsion at the ix. The struaste consisgures 5, 10AASC, the one, and the
C(GFA) bindneral and p
a more comA) hardened
a larger nustructure fo
ardening andat every sta
and its structGFA containsz and mulliten alkali solutslightly incre
alcalin
ion procesand structurdened pas
ders chemical a
process of AAmixing of GUn. Because Gi solution to
O2, and Al2Owith that of t30% GUS inby GGBFS
ent of alkali and dissolutiodening and ttime. In a
ount increaserticles, and
nt of this proce mechanic
n strength witrocess, the Gation of hydthe later st paste formhe developm) up to 360
erface betwed quartz sa20, 31] haveg, dense,
ones in comemical reacties seem to ng interfacias developin
me causes awhich is be
GUS particlehigh рН o
ucture of AAsts of threa), including
developinGUS particle
ders polyphase coplicated formd paste withumber of eleormation prod the particiage dependsure. s crystalline e), when mtion, the alka
eases. This in
11
ss featuresural elementtes
activity, GUSASBC (GUS
US-AASC witGUS does noGGBFS rati
O3) the ratithe referencncreases thfrom 5% tmetal oxid
on of GGBFSthe conditionaddition, thes the surfac
the high pHcess. Clearlyally activateth the mineraGUS particledrates in thtages of thation. This i
ment strengtdays (Figur
en the AASCand in alkale showed thand uniform
mparison witions [21] an
indicate thal transitionang interfacian increase iecause of
es caused bof the AASCASBC (GUSee structurag the minerag interfaciaes.
omposition omation of thh a structurements. GFAocess at apation of th
s on the GFA
material (22ixing AASBCali solution tntensifies the
13
s, ts
S S) h ot o o e e o e S s e e H y, d al
es e e s h
re
C i-e m h d e al al n a
by C S) al al al
of e
re A
all e A
–C o e
114 N.R. Rakhimova / Properties and microstructural characteristics of alkali-activated slag-blended cements
a) b) c) Fig. 10 - Structure types and structural elements of the AASBC hardened paste with physically active (a), physically active and reactive
blending materials (b), and reactive supplements (c). 1 is GUS, 2 and 5 are the developing interfacial transitional zones, 3 is the mineral matrix of AASC, 4 is the crystalline GFA particle, 6 is the amorphous GFA particle, 7 is the interpenetrating interfacial transitional zone, and 8 is the skeleton.
gradual hydration and dissolution of GFA in the early stages of hardening and the continuation of this process during the time with GGBFS and GFA particles of larger sizes. GFA particles of sizes >5 mm, together with the crystalline GFA grains can act as nucleation sites for the reaction products. During the time, there is gradual loosening of the surface of the GFA particles, resulting in the erosion of the interfacial transitional zone, expansion of interpenetrating constituents, including the GGBFS and GFA-based hydration products. Corrosion of the surface of GFA particles after one day of hydration and hydration products at the edge around the GFA particles containing amorphous alkaline aluminosilicate hydrates have been observed before [32, 33]. Hence, the interfacial transitional zone between the AASC fresh paste and GFA particles of glass structure can be called as "interpenetrating.” The slower reaction with the alkali component in comparison with GGBFS and the strengthening over time of the two types of interfacial transitional zones (developing and interpenetrating) create favourable conditions for strength increase of AASBC(GFA) binder at long periods of time (Fig. 3b).
GFA particles with sizes < 5 µm exhibit high reactivity even for early stages of AASBC (GFA) binder hardening. It is assumed that the amorphous silica dissolved from the GFA glassy phase removes the Са2+ from the GGBFS grains. The extraction of Са2+ from GGBFS to the solid phase shifts the chemical equilibrium between the oxides in the direction of maintenance of a high concentration of Na2O, resulting in the latter continuing to activate the GGBFS to achieve the equilibrium conditions of the component concentrations, typical for reference AASC. In addition, as a result of the cation exchange 2Nа+ → Са2+ in the liquid medium, caustic alkali is formed. There is a strong possibility that this alkali reacts with amorphous silica with sodium silicate formation, the anionic constituent of which is similar to the primary products of dissolution in the alkali aluminosilicate source, which serves as their
additional reserve. Krivenko [34] has observed that the chemical activity of GGBFS is defined by the amount of vitreous phase and рН and also the presence of the additives, which are capable of reacting with the hydrolysis products of the GGBFS glass and forming a crystalline skeleton of the hardened cement hardened paste. It is probable that GFA increases the reactivity of GGBFS. The introduction of GFA may lead to the additional activation of GGBFS, increasing the concentration of dissolution products of the starting materials and the volume of low-basic calcium silicate hydrates at the early stages of the hardening process, which reinforces the mineral matrix of AASC hardened paste and results in increased compressive and bending strengths (Table 3). Thus, the structure of the GFA-AASC hardened paste consists of three structural elements (Figure 10b) including, the mineral matrix with reduced basicity of the hydration products [21], a developing interfacial transitional zone, an interpenetrating interfacial transitional zone, crystalline, and amorphous FA particles. Figure 11a shows partly reacted fly ash particles in the volume of the binder gel, interfacial zone. 3.4.3. MS-AASC binders
МS shows extremely high reactivity, showing reactivity from early stages of hardening of AASBC (MS) binders. The MS action mechanism is analogous with those of the FA fine particles consisting of aluminosilicate glass (Figs. 6–9). The reactive MS binds with Ca2+ solubilised from GGBFS in the calcium silicate hydrates [30], forming a reinforcing skeleton of AASBC(MS) hardened paste. Figures 11b,c show that this skeleton penetrates the binder gel. Hence, the AASBC(MS) binders show the highest compressive and bending strength characteristics of AASBCs from an early stage of hardening.
The structure of AASBC (MS) hardened paste consists of structural elements (Figure 10c) including the mineral matrix with reduced basicity of the hydration products and the skeleton.
1 2 3 45 6 7 8 8
N. R. Rakhimova / Proprietăţile şi caracteristicile microstructurale ale unor lianţi micşti cu zgură, activaţi alcalin 115
a
b
c
Fig. 11 - Scanning electron images of the hardened AASBC
hardened pastes a) and b) AASBC (GFA), and c) AASBC (MS).
4. Conclusions The results of the study are summarized in
the following conclusions. 1. The effect of the siliceous mineral
supplementary materials on the properties of fresh and hardened AASBCs properties depends on the activity, content and fineness of blending materials (GUS, GFA, MS), GGBFS basicity, curing conditions, and pre-hardening time.
2. The admixtures GUS, GFA (class F), and MS are effective for strength improvement of AASBC based on neutral and acid GGBFS, activated with sodium carbonate solution. The effectiveness factor values for GUS, GFA, and MS are 0.9–1.14, 0.9–1.6, and 1.9, respectively.
3. GUS and GFA are also effective for GGBFS replacement up to 50% in AASBC. Appropriate Ssp of GUS for AASBC is higher than 500 m2/kg and the upper limit of Ssp of GFA is 500 m2/kg.
4. The influencing factors affect the strength of the AASBC in the following order:
- When introducing GUS - Ssp of GUS > (GGBFS basicity, curing conditions, and pre-hardening time).
- When introducing GFA - (Ssp of GFA, curing conditions, GGBFS basicity) > the pre-hardening time.
- When introducing MS - curing conditions > basicity of GGBFS > pre-hardening time.
5. The structural elements of AASBCs hardened pastes are as follows:
- With physically active blending materials (i.e. GUS), the elements are the mineral matrix of AASC, particles of blending material, and the developing interfacial transitional zone.
- With physically active and reactive blending materials (i.e. GFA), the elements are the mineral matrix of AASC, particles of crystal and amorphous structure of the blending material, developing and interpenetrating interfacial transitional zones, and skeleton.
- With reactive admixtures (i.e. MS), the elements include the mineral matrix of AASC and the skeleton.
REFERENCES
1. K.L. Scrivener, A. Nonat, Hydration of cementitious materials, present and future, Cement and Concrete Research, 2011, 41, 651.
2. N.R. Rakhimova, R.Z. Rakhimov, A review on alkali-activated slag cements incorporated with supplementary materials, Journal of Sustainable Cement-Based Materials, 2014, 3(1), 61.
3. B. Lothenbach, K.L. Scrivener, R.D.Hooton, Supplementary cementitious materials, Cement and Concrete Research, 2011, 41, 1244.
4. M.A. Smith, G.J. Osborne, BFS/Fly ash cements, Journal of World Cement Technology, 1977, 8(4), 223.
5. M. Tang, Optimum mix design for alkali-activated slag-high calcium fly ash concrete, Journal of Shenyang Architecture and Civil Engineering Institute, 1994, 10(4), 315.
116 N.R. Rakhimova / Properties and microstructural characteristics of alkali-activated slag-blended cements
6. F. Puertas, S. Martı́nez-Ramı́rez, S. Alonso, Alkali-activated fly ash/slag cements: Strength behaviour and hydration products, Cement and Concrete Research, 2000, 30(10), 1625.
7. C. Lu, The research and the reactive products and mineral phase for FKJ cementitious material, In Proceedings of 9th International Congress on the Chemistry of Cement,New Delhi, India, III, 1992, 319–324.
8. D. Li, J. Shen, Y. Chen, et al.,Study of properties on fly ash–slag complex cement, Cement and Concrete Research, 2000, 30(9), 1381–1387
9. C. Lu, The preliminary research of the fly ash-slag-alkali concrete, in Proceedings of 2nd Beijing International Symposium on Cements and Concrete, Beijing, 1989, 2, 232.
10. J.L. Provis, R.J. Myers, C.E. White et al., X-ray microtomography shows pore structure and tortuosity in alkali-activated binders, Cement and Concrete Research, 2012, 42(6), 855.
11. T. Yang, X. Yao, Z. Zhang et al., Mechanical property and structure of alkali-activated fly ash and slag blends, Journal of Sustainable Cement-Based Materials, 2012, 1(4), 167.
12. N. Marjanović, M. Komljenović, Z. Baščarević, et al. Physical–mechanical and microstructural properties of alkali-activated fly ash–blast furnace slag blends, Ceramics International, 2015, 41(1), 1421.
13. C. Shi, R.L. Day, Early strength development and hydration of alkali-activated blast furnace slag/fly ash blends, J. Adv. in Cem. Res. 1999, 11(4), 189.
14. F. Puertas, A. Fernandez-Jimenez, Mineralogical and microstructural characterization of alkali-activated fly ash/slag pastes, J. Cem. and Concr. Compos., 2003, 25, 287.
15. S. Puligilla, P. Mondal, Role of slag in microstructural development and hardening of fly ash-slag geopolymer, Cement and Concrete Research, 2013, 43, 70.
16. T. Yang, X. Yao, Z. Zhang et al., Mechanical property and structure of alkali-activated fly ash and slag blends, Journal of Sustainable Cement-Based Materials, 2012, 1(4), 167.
17. N.R.Rakhimova, R.Z.Rakhimov, Properties and structure formation of a stone of compositional slag alkaline bindings with siliceous mineral additives In Proceedings of the XIII International Congress on the Chemistry of cement, Madrid, July 2011, edited by Á. Palomo, A. Zaragoza, J.C.L. Agüí, p.200.
18. N.R.Rakhimova, R.Z.Rakhimov, Properties of alkali-activated slag cements, Zement-Kalk-Gips International, 2012, 11, 32.
19. V.D. Glukhovsky. Slag-alkali Cements and Concretes, Vysscha Shkola Publisher, Kiev, 1978.
20. C. Shi, P. Xie, Interface between cement paste and quartz sand in alkali-activated slag mortars, Cement and Concrete Research, 1998, 28, 887.
21. J. Deja, Carbonation aspects of alkali activated slag mortars and concretes, J.Silicates Industriels, 2002, 67, 37.
22. H. Xu, J.L. Provis, J.S.J. Van Deventer, P.V. Krivenko, Characterization of aged slag concretes, ACI Mater. J., 2008, 105, 131.
23. N.R. Rakhimova, Dr thesis, Alkali-activated slag cements and concretes with silica and silica-alumina additives, Kazan State University of architecture and engineering, Russian Federation, 2010.
24. J. Benezet, A. Benhassaine, The influence of particle size on the pozzolanic reactivity of quartz powder, Powder Technology, 1999, 103, 26.
25. V.D. Glukhovsky, P.V. Krivenko, P.V. Starchuk, Alkali-activated slag concretes based on fine-grained aggregates, Higher school, Kiev, 1981.
26. V.V. Gerasimov, Inorganic polymers based on silicon and phosphorus oxides, Stroiizdat, Moscow, 1993.
27. R. Аiler, Chemistry of silica, World, Moscow, 1982. 28. V.I. Korneev, Manufacture and application of water glass,
Stroiizdat, Leningrad, 1991. 29. V.B. Tolstoguzov, Inorganic polymers, Science, Moscow,
1967. 30. V.A. Korolev, Binder, USSR Author's certificate No.161516,
1990. 31. V.P. Ilyin In Proceedings of the 1st International Conference
on Alkaline Cements and Concretes, Kiev, 1994. 32. C. Shi, R.L. Day, Early strength development and hydration
of alkali-activated blast furnace slag/fly ash blends, Advances in Cement Research, 1999, 11(4), 189.
33. F. Puertas, A. Fernandez-Jimenez, Mineralogical and microstructural characterization of alkali-activated fly ash/slag pastes, Cement and Concrete Composites, 2003, 25, 287.
34. P.V. Krivenko, Mechanisms of the structure and property formation process of alkali-activated cement stone, Cement (in Russian), 1985, 3, 13.
****************************************************************************************************************************